72 Transportation Research Record 843 Dune Sand-Aggregate Mixes and Dune Sand-Sulfur Mixes for Asphalt Concrete Pavements
M.N. FATANI AND H.A. SULTAN
Results are presented of a study to determine the feasibility of using dune sand Tanner Pit southeast of Tucson, Arizona. The maxi in asphalt concrete pavement in hot, desertlike climates through the use of mum particle size is 3/4 in, with only 2 percent re one-size crushed aggregates, dense-graded aggregates, and powdered sulfur in tained on a 3/8-in sieve, as shown in Figure 1. The the sand-asphalt mixes. Engineering properties, including Marshall design specific gravity is 2.65, the Los Angeles abrasion parameters, compressive strength, tensile strength, modulus of rupture, and dy loss (AASHTO T96-65) is 20 percent, and the sodium namic modulus of elasticity, of the various mixes are given and discussed. sulfate soundness test (AASHTO Tl04-65) gave a loss Results indicate that a mixture of dune sand and asphalt is weak, unstable, of 6 percent. More than 90 percent of this gravel easily deformed under light loads, and therefore unacceptable for pavement had at least one mechanically fractured face. construction in a desertlike environment. Introducing a one-size crushed gravel at various ratios improved the mix but not sufficiently for it to pass the required standards. Introducing a dense-graded aggregate to the mix Dense-Graded Aggregate raised Its properties to the required standards; however, such improvement was only achieved when the ratio of dense aggregate to dune sand was at The dense-graded aggregate is a mixture of four least 60 percent to 40 percent. The results also demonstrate that the use of materials: 15 percent 3/4-in crushed gravel, 25 powdered sulfur In the sand-asphalt mix•• rur.lui:•• lhe ul-'limum a~phalt con· percent 3/8-in crushed gravel, 53 percent Pcu1La110 tent, considerably increases the qualities of the mix even under severe en Wash sand, and 7 percent fly ash. vironmental conditions, and reduces the pavement thickness required to protect the subgrade from deflections and the surface layer from fatigue 1. Crushed gravel, 3/4-in: This is the one-size cracking. A tentative thickness design chart of a pavement for typical traffic crushed gravel discussed above. and environmental conditions found in desertlike areas is also presented. 2. Crushed gravel, 3/8-in: This material, as shown in Figure 1, has a maximum size of 3/8 in with In Saudi Arabia an extensive road-building program about 20 percent retained on a No. 4 sieve. The is currently under way. During the second five-year specific gravity is 2.63, the Los Angeles abrasion plan (which ended in 1980), about 7460 miles of as loss is 20 percent, and the sodium sulfate soundness phalt roads and 6200 miles of rural roads were com pleted (,!). Although a good percentage of these test loss is 6 percent. More than 90 percent of roads and of others currently planned go through this gravel had at least one mechanically fractured sand-dune areas, dune sand had not been used as a face. pavement material. Instead, good-quality aggre 3. Pantano Wash sand: This sand was obtained gates, which are in scarce supply, are being im from the Pantano Wash, a dry river in t he Tucson ported from other localities at a considerably high area. The unwashed sand had a sand equivalent cost. Therefore, it seems appropriate to consider (AASHTO Tl 76-6) of about 31 and 8 percent passing a the potential of large-scale use of dune sand as a No. 200 sieve. The sand used in this investigation pavement material to reduce cost and save the good was washed thoroughly until all the fines were quality aggregates for other construction uses. washed out. The clean sand gave a sand equivalent Dune sand is a natural material that exists in of 83. The grain-size distribution for the washed sand is given in Figur~ 1. abundant quantities on every continent and under al 4. Fly ash: The fly ash (Navajo fly ash) used most every climatic condition. Although the exis in this investigation was obtained from the Western tence of dune sand is characteristic of desert Ash Company. Its major constituents include 52. 7 areas, sand is also commonly found along the shores of seas, lakes, and large streams (2). percent silicon dioxide, 20. 5 percent aluminum This paper presents some data from a comprehen oxide, and 4.9 percent ferric oxide · The grain sive investigation aimed at determining the feasi size distribution is given in Figure 1. bility of using dune sand in asphalt concrete pave ment through the use of one-size crushed aggregates, These four materials were combined in the propor a dense-graded aggregate, and powdered sulfur in the tions indicated above to form a dense-graded aggre sand-asphalt mixes (]) • gate that met the Asphalt Institute gradation limits C~.• Specification IVb) • MATERIALS USED Asphalt Dune Sand The asphalt cement used throughout this investiga The dune sand selected was obtained from the dunes tion was an AR-4000 (6-70 penetration), which is widely used for hot mixes in road construction in area west of Yuma, Arizona. The sand is composed of Arizona. Its physical properties are given else subrounded to subangular grains with very fine tex ture. It is essentially a one-size material with a where (,!,],). uniformity coefficient of about 2 and a specific gravity of 2.65. The grain-size distribution for this sand is included in Figure 1. Quartz is the major constituent of this material. Other minerals The sulfur used in preparing the sand-asphalt-sulfur such as plagioclase feldspars, orthoclase feldspars, mixes was a bright yellow elemental sulfur in a pow and micas are also present. It is classified as an dered form that had 99. 5 percent purity. It is a A-3(o) material. commercial-grade sulfur known as Ortho Sulfur and manufactured by the Chevron Ortho-Division Company. One-Size Crushed Gravel Additional data on this sulfur are given elsewhere (],). Th is aggregate is primarily a mixture of crushed gravel and limestone and was obtained from the Transportation Research Record 843 73
Figure 1. Grain-size distribution curves for aggregates used. Additional mixing at medium speed for about 60 s was required to coat the particles uniformly with the SIEVE NUMBER AND SIZE asphalt. This combination was designated Test IOO llO I S/8" 514" too '° .. 4 112· 1· Series C. IOO 0 For the (A-S) mix, the heated asphalt was added 11 !ill : ~:::,hfrrt '·•.Ji•• /, I _t:Jr Ao h_.:;;~.; -,,, 1111 111 1 I~ I I to the preheated sand and the two components were 1 1 1 I I 10 I 1! 1 '111 ~ mixed together in the preheated bowl of the Hobart I _v.,,,.~.. - 20 C-10 mixer for 60 s. The sulfur (at room tempera I ture) was then added and the entire batch was re : iY,{~H~ I 30 0 mixed for an additional 60 s. This combination was _ WH f'lelS I I Po. .nrono" ~ z designated Test Series D. 4C "'- I - Saod ~ ~ - I I I ~ After mixing, both mixtures were placed in an I ! I ~ ... .• I llO "'....a: oven at 280 ± 5°F until compaction. I- ' I 11--319•_ 314•_ z ~ ~ Cro 1h 1 d Crr.ithtd : - ~ ' Gf Test Series c and D and 15 percent sulfur) had been compacted by using 2, 5, 10, 15, 25, and 50 blows per face (3). All specimens used in these series were from mixes The Marshall stability results for- the (S-A) that had an asphalt content of 5 percent by total mixes and the (A-S) mixes are given in Figure 4 with weight. This was based on various reported results data for a similar mix that used a binder of molten on mixes of sand, asphalt, and sulfur (7-9). The sulfur and asphalt added to the same dune sand in a sulfur content in the mixes varied from zero to 20 one-wet-mixing process (7). The results indicate percent by total weight. that both the (S-A) and (A-S) mixes have their maxi For the Marshall stability test on these mixes, mum Marshall stability values at 15 percent powdered each specimen was compacted by applying 10 blows to sulfur. The maximum Marshall stability value for each end of the specimen. It was decided this would the (A-S) mix was 6500 lbf and that for the (S-A) be the optimum compaction effort after a series of mix was 5983 lbf. A sulfur content greater than 15 test specimens (80 percent sand, 5 percent asphalt, percent reduced the stability of both mixes. The sharp increase in Marshall stability from a weak, unstable sand and asphalt mix (82 lbf) to a strong, stable mix of sand, asphalt, and sulfur is Figure 2. Relation between Marshall stability and ratios of dune sand to attributed to the filling of the voids by the re one-size crushed gravel (Test Series A). crystallized sulfur. The sulfur acts as a filler in I addition to bonding the sand particles and thus in creasing their interlocking. The (S-A) mix gave a slightly lower stability value because the particle 300 One Size Crushed Grovel- - • - 5 % Asphalt Content to-particle contact was asphalt, a viscous material, Dune Sand-Asphalt Miit.es -•- 6% rather than sulfur, a solid material below 235°F. -•-1% Aboaziza (7) used a premixed blend of molten sul fur and asphalt mixed in a high-shear mixer at 300°F 250 as the binder, and he mixed it with the dune sand in :!! a one-wet cycle. Similar results to those shown in ,: Figure 4 were reported but with slightly lower Mar r :J 200 shall stability values. iii Figure 5 presents the effect of increasing the ~ VJ percentage of sulfur in the mix on its unit weight. ...J The increase of sulfur between 0 and 15 percent sig Figure 3. Relation between Marshall stability and ratios of dune sand to dense Figure 4. Relation between Marshall stability and sulfur content in mixtures of graded aggregate (Test Series B). dune sand, asphalt, and sulfur. ' • Asphalt-Cooled - x - 4% A1phal t Content Sond-Sulrur (A-S)y• 2400 -•-5% A Sulfur - Cooled ~ -•-6% 6 000 Sond-Aspholl (S-A) A * Molten Sulfu•·A•phollN/ ~ ~~:c:,•: Mixin9 --J... / ~ • 2000 Dense Graded-Dune Sand- AsphoH Mi us 5000 Ab~ ' '" (l981 ) \\ :!! ,: :!! r ,.: :::; 1600 1-- iii :i 4 0 00 r Vl"" 1--"' ...J ""Vl 100/0 80120 60/40 40/ 60 20180 0/100 J5 10 15 20 OUNE SAND - DENSE-GRADED AGGREGATE RATIO SULFUR CONTENT, % by Totol Weigh! Transportation Research Record 843 75 the brittle nature of the recrystallized sulfur and of sand, asphalt, and sulfur is much lower than that the increased unit weight of the mix. of conventional asphalt concrete mixes for a given Figure 7 shows the effect of the sulfur addition content of air voids (10). Burgess and Deme (10) on the percentage of air voids in the mix, which in reported that at 15 percent content of air voids, dicates sharp reductions of the air voids and re sand-asphalt-sulfur mixes were considered impervious sults in more stable mixes. The mix of dune sand with a permeability coefficient of about 1.0 x and asphalt without sulfur has a very high amount of 10" 8 cm•. Conventional dense-graded asphalt con air voids due to the uniformity of the sand-grain crete mixes reached similar permeability values at size. When sulfur is added to the mix, it fills the about 4-5 percent content of air voids (!.Q) • voids and results in a denser mix on recrystalliza Similar results were obtained by Aboaziza Ill by tion. It may be pointed out that although the per using molten sulfur. These results are also given centage of air voids in the mixes exceeds the recom in Figures 5 through 7 for purposes of comparison. mended range for the Marshall design criteria (6), Based on the above test results, the (A-S) mixes it had been shown that the permeability of mixtures in Test Series D appeared to give the overall best performance for the mixes of sand, asphalt, and sul fur. The (A-S) mix with dune sand, asphalt, and sulfur contents of 80, 5, and 15 percent, respec Figure 5. Relation between unit weight and sulfur content of mixes of dune tively, was therefore considered for further test sand, asphalt, and sulfur. ing. Additional tests included tensile strength, compressive strength, soaked compressive strength, flexural strength, and dynamic modulus of elasticity. 130 • Asphalt-Coated So"d Sulfo r ~ Tensile-Strength Test 4 .A Sulfur-Cooled S ond · A1pt10 1 ~_ /J.~• * Molten Sulfur-Asphotl /..~ ..;.. .&. The static double-punch test (11) was proposed as a One Wei Mmng P r1~cess f simple indirect tension test ~or determining the ",,. .. __ ,,.I tensile strength of concrete. Fang and Chen (Jl) developed, both theoretically and experimentally, the applications of the double-punch test to cohe sive soils. Jimenez (13) extended the use of the double-punch method to test asphaltic mixtures for tensile strength, indicating its better repeatabil ity than the split-cylinder (Brazilian) test. ,_ '"""' '"""' [ / :.: 2 Mucmo The static double-punch test is conducted by ~ Mtlhod ~,. ~'"'"Asphalt Contenl, !5 % using two steel disks (punches) centered on both Number of Blows per Foc:e, 10 flat ends of a cylindrical specimen. The vertical 11 0~ load is then applied slowly on the punches until the * specimen reaches failure. The tensile strength of the specimen is calculated from the maximum load by 105 a simple equation based on the theory of perfect plasticity (12). In this investigation, specimens for the static double-punch test were 4 in in diameter and 2. 5 in 0 5 10 15 20 high. The mixes of sand, asphalt, and sulfur were SULFUR CONTENT, % by Toto I Weight compacted by the Marshall compactor (10 blows per Figure 6. Relation between Marshall flow and sulfur content of mixes of dune Figure 7. Relation between percentage of air voids and sulfur content of mixes sand, asphalt, and sulfur. of dune sand, asphalt, and sulfur. Variable' I Sulfur Content 2. Miaing Method Aspholl Cionttnt. ! % Number of Blows per Face, 10 30~\.. - ~ • Aspholt-Cooltd Sond·Sulfur V- • Sulfur- Cooled Sand-Asphalt .E ~ ../ * Molten Sulfur-Asphalt One W1! Mi~ing Process ~ 20 ~ Aboazlza (1911) ui 0 0 > a: :;;: 15 Variable: I Sulfur Content 2 Mixin4'1 Method 10 Atpholl Content, 5 % Number of Blows pe r Fo ce, 10 o'--~~....L..~~~'--~~....L..~~~"--~~-' 0 5 10 15 20 0 10 15 SULFUR CONTENT, % by Toto I Weight SULFUR CONTENT, % by Totol Weight 76 Transportation Research Record 843 Figure 8. Effect of temperature on tensile strength of various mixes. phalt mix on addition of 15 percent sulfur in either molten or powdered form. The mix without sulfur had very low tensile-strength values at temperatures • Sulfi..ir: A1pha lt 0 Sand above 70°F. Similar tensile-strength values were .. 15 5 80 (Powder Sulfur) • 15 5 ' 80 (Molten Sulfur) obtained for the powdered-sulfur (A-S) mix and the 200 • 0 65 ; 935 (No Sulfur) molten-sulfur (one-wet-cycle) mix, with significant reductions in strength values as the curing (and testing) temperature increased from 40°F to 140°F, This reduction in tensile strength is attributed to the fact that the binder, and thus the mix, becomes · ~ 150 • softer and more flexible, which results in a pro ...... : gressive decrease in tensile strength as temperature "'z increases. ...."'a: en Unconfined-Compression Test .J 100 "'(;; z The compressive strength of the (A-S) mix was deter "'.... mined according to the ASTM,01074 (AASHTO Tl67) test procedure. The specimens were compacted statically to their respective unit weights obtained in the 50 Marshall test. The specimens were 4 in in diameter and 4 in in height. After compaction, the specimens were cured at various temperatures for 24 h prior to testing. After being cured, the s1,1eclme11s were im mediately tested in compression at a deformation rate of 0.05 in/min. Figure 9 shows the compressive-strength results TEMPERATURE, "F for the (A-S) mix as a function of curing (and test ing) temperatures, which indicates that the compres sive-str ength values decreased as the temperature Figure 9. Effect of temperature on compressive and immersion compressive increased due to the softening effects of the strength of mixes of dune sand, asphalt, and sulfur. binder. It may also be pointed out that the addition of sulfur to the sand-asphalt mix significantly in creased the compressive strength ·;;; Q. Immersion Compressive Strength ....£ 750 ~ The immersion compressive strength of specimens similar to those tested for the standard compressive ...."'a: en strength was tested after an additional 24-h immer "'> sion in water at the same temperatures at which they ~ 500 .... were cured. The immersion compression test results a: at the three temperature levels, shown in Figure 9, Cl. :I: are lower than the respective standard compressive 0 u strength values. If we take the ratio of the immersed compressive 250 strength to the respective standard compressive strength as an index for retained strength, the re sults indicate a minimum retained strength of about 73 percent for the temperature range tested. The highest index of retained strength for the sand-as phalt mix (without sulfur) was reported by Aboaziza TEMPERATURE, "F (7) to be only 45 percent. This demonstrates that the addition of sulfur in sand-asphalt mixes sig nificantly increases their strength and durability and hence their resistance to stripping or debonding. end). Specimens were tested in an Instron Universal Testing Machine at a deformation rate of 0.05 Flexural Strength Test in/min. After compaction, the specimens were cured at various temperatures for 24 h prior to being The modulus of rupture, which measures the flexural tested. The curing (and testing) temperatures of strength of the (A-S) mixes, was determined by using 40 .±. 5°F, 77 .±. 5°F, and 140 .±.. 5°F were selected to simple beams with third-point loading tests. These represent the average low temperature, the average tests were conducted according to the ASTM C78 mean temperature, and the average high temperature (AASHTO T97) test procedure. The beam specimens, in a desert environment, respectively <1,14). 3 x 3 x 11. 25 in, were statically compacted to the Tensile-strength data for these (A-S) mixes are corresponding unit weights obtained in the Marshall given in Figure 8 along with strength data for a test. After compaction, the beams were cured at the sand-asphalt mix without sulfur and data for a mix various temperatures for 24 h. After being cured, of sand, asphalt, and molten sulfur obtained by the beams were tested at a deformation rate of 0.05 Aboaziza Figure 10. Effect of temperature on flexural strength of various mixes. given in Figure 10 along with strength data for a sand-asphalt mix without sulfur and data for a mix of sand, asphalt, and molten sulfur obtained by Sulfur' Asphalt= Sand Aboaziza <1>· The results indicate significant im "' 15 5 '80 (Powder Sulfur) provement in the flexural strength on addition of 600 • 15 5 ' 80 (Mollon Sulfur) • 0' 65 '935 (NoSullur) sulfur, in either powdered or molten form, to the sand-asphalt mix. It may be pointed out that at 77°F, the mix without sulfur had no flexural 500 strength and could not support its own weight, ·;; whereas with 15 percent sulfur, the mixes of sand, c. ...; asphalt, and sulfur had a flexural strength of about a: 200 psi at that temperature. ,_::> 400 Q. ::> a: Dynamic Modulus of Elasticity "- 0 Molten Sulfur VJ 300 ::> One Wet Mixino Proc11s CL) The dynamic or resilient modulus of elasticity is ...J ::> considered one of the most important characteristics 0 0 of a pavement material. Use of the elastic theory :::!! 200 (multilayered systems) is essentially dependent on application of this modulus. The dynamic modulus of elasticity is defined as the ratio of applied stress to the recoverable strain as obtained by dynamic 100 measurements. In this investigation, the dynamic 0% Sulfu~r CV • modulus of elasticity for the (A-S) mixes was ob tained by using the dynamic double-punch test (j2). 0 Jimenez (15) extended the use of the double-punch 75 25 50 "· 100 125 150 test to dete;mine the dynamic modulus Eo of as TEMPERATURE, 'F phaltic concrete. Fatani Cll, Aboaziza (ll, and Jimenez (15) have discussed this procedure and the theory used to formulate the equation for calculat Figure 11. Effect of temperature on dynamic modulus of elasticity of various ing En in more detail. mixes. The test data for dynamic modulus E0-values for the (A-S) mixes are given in Figure 11 along with data for a sand-asphalt mix without sulfur and data Sulfur Asphalt Sand for a mix of sand, asphalt, and molten sulfur ob A 15 5 = 80 (Powder Sulfur) tained by Aboaziza (7) • The data indicate the 12 • 15 I 5 80 (Molten surrur) • 0 6 5 ' 93 5 CNo Sulfur! superiority of the mixes of asphalt, sand, and sul fur over the sand-asphalt mixes (without sulfur), particularly at high temperatures. All mixes indi ·;; 10 cate a reduction in Ee-values with increases in c. temperature. "'2 x Table 1 shows a comparison between the values of 0 the dynamic modulus of elasticity at the three tem "',,; 8 perature ranges reported here and those from other ::> ...J ::> One Wei Mixino Process investigations <1• 16-l!!.) by using different tech Cl Cl) 0 niques with or without sulfur. It can be noted that :::!! 6 u the dynamic modulus results from the double-punch i test are in general agreement with those obtained by z using the Schmidt method (16), especially at the low >"' Cl testing temperature. The weak, unstable sand-as phalt mix gave En-values either below or at the lower range of the typical values of dynamic modulus obtained for a dense-graded aggregate-asphalt mix along the entire temperature range under considera tlun (18). At low temperatures, the addition of sulfur to o~~~~~~~~~~~~~~~~~-' sand-asphalt mixes improved the En-values to the 25 50 75 100 125 150 middle ranges of En for the dense-graded aggre TEMPERATURE, •f gate-asphalt mix. The improvement approached the Table 1. Comparison of dynamic moduli of elasticity 5 obtained by different methods. E0 (psi x 10 ) Asphalt Sulfur Method Aggregate (%) (%) 40"F 70°F IOO"F Reference Schmidt Beach sand 6 13.58 8.9 4.4 2.3 (16) Schmidt Beach sand 4 13.58 10.5 11.9 3.8 cm Deflectometer Dune sand 6 0 10.0 1.5 0.3 (17) Double punch Dune sand 6.5 0 9.0 3.0 0.0 (f) Double punch Dune sand 5 15 8 11.2 7.1 4.6 Cll Double punch Dune sand 5 15b I I.I 7.0 4.0 C!) Typical values Dense mix (J!) Mean 0 16.0 5.0 1.0 Range 0 9-27 4-9 0.7-2.2 8 Molten sulfllr. bPowdered sulfur. 78 Transportation Research Record 843 higher ranges of Eo at intermediate temperatures, PRELIMINARY THICKNESS DESIGN and it exceeded the given ranges at the high tem peratures (which are critical for arid and semiarid A preliminary theoretical analysis by using the climates). elastic-layered system was conducted to determine the thickness requirements for selected mixes of as phalt, sand, and sulfur under typical loading and subgrade conditions (1). The thickness design re quirements were evaluated for a 9000-lbf wheel load Figure 12. Design chart (18 000-lbf single-axle load) for mix of dune sand, (18-kip axle load) i this load was applied by dual asphalt, and sulfur. wheels (each 4500 lbf). The contact pressure was 10• assumed to be 80 psi acting on a circular area of :::t:= 4.23-in radius. The center-to-center spacing of the _.).._ wheels in the dual axle was also assumed to be three __ ,_ "it times the radius of the loaded area, i.e., 12.69 - ... ='/- - - in. The Chevron Computer Program (19) with a slight !: 101 ' modification that permits the use of dual-wheel loading (17) was used to calculate the developed a: "'::> _J I stresses and strains at various strategic locations I " I in the pavement section. The computer program list ~ rt- - ·- _.J_ - _._ .... -r /1 I - ing and description are given elsewhere (~_,.!ll. < IOT 1=.::--0~. -gJ, Ulz !::;-.":l 9-~ §:· Based on data reported by others (.!l,±.Ql , an 0 ..,-- \) ... ,o.. ·' average dynamic modulus for dune-sand subgrades is ~ r---r- i-·· ::' u I I "'~ ...- about 3 x 10' psi. The E0-values used for the ::; --r- -- mixes of sand, asphall, aud sul[UL in tills ctmilysls a. --,- !/- - - -I ·- ~ 10• at 40, 77, and 140°F are 1.11 x 10 6 , 6.3 x 10 5 , 0 and 2.4 x 10 5 psi, respectively. < f--'-1 9 ' Based on the above design criteria and the elas u. f I I tic properties of the dune sand and (A-S) mixes, a 0 I ~ - --I'- - 1--' ~ 1 0~ "jJ-,- set of thickness design charts was developed (1) • OJ The outline of the development procedure for the :f ::> charts is detailed elsewhere (3). Figure 12 pre z sents a thickness design chart -for the (A-S) mixes I I I I ' I I _ J. _ I 1..-- at these three effective temperatures with a sub -,- I- I -+ 10• t ·H grade modulus of 3 x 10' psi [California bearing ratio (CBR) of 20]. As shown, the design curves are steep and closely spaced, which indicates that a 4 s e 10 20 40 so eo 100 PAVEMENT THICKNESS, in. great increase in the expected number of load appli cations requires only a small increase in pavement thickness and that this type of pavement can with stand a great number of load repetitions with rea Figure 13. Design chart (18 000-lbf single-axle load) for mix of dune sand sonable thickness requirements, even at high tem and asphalt. peratures. AlSalloum 1.!ll presented a similar design chart 2 for sand-asphalt mixes (without sulfur), which is reproduced here as Figure 13. If we compare Figures SUBGRADE MODULUS 1 E2 • 30,000pol I / 12 and 13, there is an indication that significant CCBR • 20) I I reductions in pavement thickness required to with / / stand the same conditions are achieved with addi tions of sulfur. It may be pointed out that the above analysis and .,o·" ' chart are of a preliminary nature and were based on '- the assumed design criteria. Therefore, they should be used as guidelines for preliminary thickness de , , , sign only until more detailed and inclusive design . , charts are developed. ~' , , , , / CONCLUSIONS , A mix of dune sand and asphalt is weak, unstable, easily deformed under light loads, and nondurable. It is not acceptable for pavement construction in a desertlike environment. The introduction of a crushed gravel at different ratios improved the mix but not to the required level. The introduction of a dense-graded aggregate resulted in some improve ment, but the quantity of costly high-quality min eral aggregate required to meet specifications made the mix uneconomical. The introduction of 15 percent sulfur to a mix of dune sand and asphalt gives the following beneficial results: 1. Reduces the optimum asphalt content of the 3 10'---~-'--~'-~-'--~'-~-'--~L...L~~-'-~...._~_,_~....__.__._...... mix from 6.4 to 5.0 percent, a reduction of 22 per 1 5 s 7 e g io 20 30 40 50 60 centi PAVEMENT THICKNESS IN INCHES 2. Considerably increases the engineering quali- Transportation Research Record 843 79 ties of the mix even under severe environmental con 7. A.H. Aboaziza. Characterization of Sulfur-As ditions (such improvements are evident in the Mar phalt-Dune Sand Paving Mixtures. Univ. of Ari shall criteria, tensile strength, compressive zona, Tucson, Ph.D. dissertation, 1981, 333 pp. strength, modulus of rupture, and dynamic modulus of 8. I. Deme. Processing of Sand-Asphalt-Sulfur elasticity values); Mixes. Proc., AAPT, Williamsburg, VA, Vol. 43, 3. Significantly reduces the pavement thickness 1974, pp. 465-490. required under similar loading and environmental 9. D.Y. Lee. Modification of Asphalt and Asphalt conditions; and Paving Mixtures by Sulfur Additives. Indus 4. Removes the need to import dense-graded ag trial and Engineering Chemistry, Product Re gregate to the site because the engineering quali search and Development, Vol. 14, No. 3, 1975, ties of the mix of dune sand, asphalt, and sulfur pp. 171-177. are equal to or better than those of conventional 10. R.A. Burgess and I. Deme. The Development of dense-graded aggregate-asphalt mixes. Not having to the Use of Sulphur in Asphalt Paving Mixes. use dense-graded aggregate in the mix further re Presented at Sulfur Use Symposium, American duces construction costs, especially if the aggre Chemical Society, Na.tional Meeting, Los gate has to be hauled a considerable distance. An Angeles, CA, 1974. additional benefit is that high-quality mineral ag 11. W.F. Chen. Double Punch Test for Tensile gregate inay be saved for other purposes and/or for Strength of Concrete. Journal of the American reducing the ecological damage caused by excavating concrete Institute, Proceedings Vol. 5 7, 1970, borrow areas and hillsides. pp. 993-995. 12. H.Y. Fang and W.F. Chen. New Method for De ACKNOWLEDGMENT termination of Tensile Strength of Soils. HRB, Highway Research Record 345, 1971, pp. 62-68. The reported investigation was conducted at the De 13. R.A. Jimenez. Testing for Debonding of Asphalt partment of Civil Engineering, University of Ari Pavements (Arizona) ••• Phase II. Arizona De zona, while M.N. Fatani was a graduate research as partment of Transportation, Phoenix, May 1975. sociate. Assistance rendered by R.A. Jimenez and 14. A.F. Bissada. Analysi s of High Temperature In A.H . Aboaziza is gratefully acknowledged. stability Failures of Heavily Trafficked As phalt Pavement. Presented at Annual Meeting, REFERENCES AAPT, Louisville, KY, 1980. 15. R.A. Jimenez. Structural Design of Asphalt 1. H.A. Sultan, M.N. Fatani, and A.H. Aboaziza. Pavements. Arizona Department of Transporta Dune Sand-Asphalt-Sulfur Mixes for Asphalt ,con tion, Phoenix, Final Rept., Nov. 1975. crete Pavements. Presented at Symposium on 16. B .M. Gallaway and D. Saylak. Tasks I: Famil Geotechnical Problems in Saudi Arabia, College iarization and Verification of Existing Tech of Engineering, Univ. of Riyadh, Saudi Arabia, nology. Texas A&M univ. Res. Foundation, Col May 11-12, 1981, 29 pp. lege Station, TX, Project RF 983, Vol. 1, 1974. 2. A.A. Rula, W.E. Graber, and R.D. Miles. Fore 17. N.M. AlSalloum. Fatigue Characteristics of As casting Trafficability of Soils, Air Photo Ap phalt Stabilized Dune Sand. Univ. of Arizona, proach. U.S. Army Corps of Engineers Waterways Tucson, Ph.D. dissertation, 1973. Experiment Station, Vicksburg, MI, Tech. Memo 18. E. J. Yoder and M.W. Witczak. Principles of 3-331, Rept. 6, 1963. Pavement Design, 2nd ed. Wiley, New York, 1975. 3. M.N. Fatani. Dune Sand Aggregate Mixes and 19. H. Warren and W.L. Dickmann. Numerical Compu Dune Sand-Sulfur Mixes for Asphaltic Concrete tation of Stresses and Strains in Multiple Pavements. Univ. of Arizona, Tucson, Ph.D. Layered Asphalt Pavement System. California Re dissertation, 1980, 319 pp. search Corporation, Richmond, CA, 1963. 4. J .c. Rosner. Let's Design Fly Ash Concretes: 20. w. Heukelom and A.G. Klomp. Dynamic Testing as Not Compare Them. Proc., Fourth International a Means of Controlling Pavements During and Af Ash Utilization Symposium, St. Louis, MO, March ter Construction. First International Con 1975, pp. 560-572. ference on Structural Design of Asphalt Pave 5. Asphalt Institute. The Asphalt Handbook. Man ments, Univ. of Michigan, Ann Arbor, 1962, pp. ual Series No. 4 (MS-4), 6th ed., College Park, 667-679. MD, 1970. 6 . Asphalt Institute. Mix Design Methods for As phalt Concrete and Other Hot-Mix Types. Manual Series No. 2 (MS-2), 4th ed., College Park, MD, Publication of this paper sponsored by Committee on Characteristics of 1974. Bituminous Paving Mixtures to Meet Structural Requirements.